Publisher’s version / Version de l'éditeur:
Food Research International, 33, August 7, pp. 617-620, 2000
READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE. https://nrc-publications.canada.ca/eng/copyright
Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la
première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à [email protected].
Questions? Contact the NRC Publications Archive team at
[email protected]. If you wish to email the authors directly, please see the first page of the publication for their contact information.
NRC Publications Archive
Archives des publications du CNRC
This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.
For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.
https://doi.org/10.1016/S0963-9969(00)00098-3
Access and use of this website and the material on it are subject to the Terms and Conditions set forth at
Membrane-based separation scheme for processing sweeteners from
stevia leaves
Zhang, Shi Qiu; Kumar, Ashwani; Kutowy, Oleh
https://publications-cnrc.canada.ca/fra/droits
L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.
NRC Publications Record / Notice d'Archives des publications de CNRC:
https://nrc-publications.canada.ca/eng/view/object/?id=dd489890-fd25-43bb-a532-54036c03e9c0 https://publications-cnrc.canada.ca/fra/voir/objet/?id=dd489890-fd25-43bb-a532-54036c03e9c0
Membrane-based separation scheme for processing
sweeteners from stevia leaves
$Shi Qiu Zhang, Ashwani Kumar *, Oleh Kutowy
Institute for Chemical Process and Environmental Technology, National Research Council of Canada,M-12 Montreal Road Campus, Ottawa, ON, K1A 0R6, Canada Received 1 October 1999; accepted 6 January 2000
Abstract
In existing processes, extraction and re®ning of glycoside based sweeteners from stevia leaves involves many process steps including extraction by organic solvents. The purpose of the present study was to develop a process of extraction and re®ning of sweeteners with reduced number of unit operations and minimization and/or elimination of chemical usage including organic solvents. It was found that water was very eective for extracting glycosides at selected pH and temperatures. It was also shown that a multi-stage membrane process was successfully able to concentrate the glycoside sweeteners. Based on the preliminary results, it appears that bitter-tasting components were washed out from the sweetener concentrate in the nano®ltration process. This work also has demonstrated that a membrane-based separation process for re®ning glycoside-based sweeteners could be viable and needs to be investigated further. Canadian Crown Copyright # 2000 Published by Elsevier Science Ltd. All rights reserved.
Keywords:Stevia; Sweeteners; Extraction; Separation; Membranes; Ultra®ltration; Nano®ltration; Micro®ltration
1. Introduction
The food industry has traditionally used sugar as the sweetening agent. However, there is increasing demand for other sweeteners partially in response to consumer preference. There are synthetic as well as natural source sweeteners in the market. A signi®cant segment of con-sumers is interested in sweeteners that are known to have nutritive values. Additionally the sweetener should ful®l the requirements relating to non-toxic nature, sugar-like taste pro®le, low calori®c value, heat and pH stability. In recent years, there has been considerable interest in stevia-based natural source sweeteners that possess many of these desired qualities. Stevia rebaudi-ana is a shrub, native to Paraguay and is grown in the Far East, South America and many other areas of the world. The sweet compounds represent about 14% constituents of dried leaves and are diterpene glycosides based on the kaurene skeleton. These are mainly com-prised of stevioside, rebaudioside A, B, D, E, dulcoside A and B (Leung & Foster, 1996). The most abundant
glycoside is stevioside followed by rebaudioside and minor components. Sweetening potencies of various diterpenes found in S. rebaudiana range from 50 to 300 while non-sweet constituents mainly are labdane diter-penes, triterditer-penes, sterols and ¯avonoids. Rebaudioside A is the most desirable component due to its sweetening potency and superior taste pro®le (Cramer & Ikan, 1986). Responding to this challenge, new cultivars with sig-ni®cantly higher concentration of Rebaudioside A have recently been reported (Brandle, 1999).
There has been considerable interest in using stevia-based sweeteners in consumer products particularly in Japan. It was claimed that stevioside had a 20% of market share of low-calorie sweeteners in Japan (Kiku-chi, 1985). According to some estimates, the market potential of these sweeteners at a penetration level of 4ÿ8% in Japan and Far Eastern countries where it is approved would be in the range of US $1200 million. There is abundant published and patent literature on the stevia re®ning from dried leaves. For example, there are more than 150 Japanese patents on the subject. The process steps essentially are extraction, pretreatment, separation and re®ning. Most of the reported processes use coagulating agents and organic solvents. Some of the selected processes utilize chromatographic separation
0963-9969/00/$ - see front matter Canadian Crown Copyright # 2000 Published by Elsevier Science Ltd. All rights reserved. P I I : S 0 9 6 3 - 9 9 6 9 ( 0 0 ) 0 0 0 9 8 - 3
www.elsevier.com/locate/foodres
$ Issued as NRCC no. 42003.
* Corresponding author. Tel.: +1-613-998-0498; fax: +1-613-941-2529.
(Matsushita & Ikushige, 1979) and chelating agents fol-lowed by solvent extraction (Kumar, 1986). A process involving pretreatment of the extract with lime and use of a series of ion exchange column is of special mention as it has eliminated the use of organic solvents (Giovanetto, 1990). However, all of these processes have complex steps and use signi®cant amounts of chemicals and/or generate sludge. There is a need of modifying this pro-cess to reduce the consumption of chemicals and waste streams. Furthermore, additional improvement in taste pro®le and colour of the ®nal product is desirable. The present work was started to address above problems by using membrane-based process with modi®cation in the extraction process step. This paper reports the eects of temperature and pH on extraction of sweeteners from dried leaves, pretreatment of the extract and subse-quently eects of temperature on separation and re®ning by a multi-stage membrane process.
2. Experimental 2.1. Feed material
Ontario-grown dried stevia leaves were supplied by Agri-Canada's Delhi Research Station. These leaves were used in extraction experiment after removing twigs and branches. Concentrations of major glucosides in leave extract as determined by HPLC analysis were dulcoside, 5.6; stevioside, 58.7; rebaudioside C, 9.8 and rebaudioside A, 87.2 mg per g of dry leaves. Other plant constituents in the leave extract were not determined. 2.2. Analytical methods
Stevia leave extract was diluted appropriately by acet-onitrile and water mixture (80:20) and allowed to stand overnight. In order to keep the chromatogram from cluttering by the peaks of ingredients other than glyco-sides, the sample for HPLC analysis was withdrawn only from the top layer. A 5 mm CSC-Sil 80 A amino column with dimensions of 2550.46 mm was used (Chromato-graphic Sciences, Montreal, Canada). Standard operat-ing conditions for HPLC analysis included a mobile phase of acetonitrile and water (80:20); ¯ow rate 1.5 ml per min and a column temperature of 28C. The cali-bration curves for stevioside, rebaudioside A and C were constructed using standards supplied by the Alberta Research Council.
2.3. Apparatus
A jacketed glass extraction column with dimensions of 60090 mm was used for column extraction while the batch extractions were done in glass beakers. Initial mem-brane evaluation and characterization for separations was
done using NRC cells (Sourirajan & Matsuura, 1985; Zhang, Kutowy, Kumar & Malcolm, 1997) as well as Sepacell (supplied by Osmonics, Minnetonka, MN, USA).
2.4. Procedure
A carefully weighed amount of dried leaves were placed in a standard glass column and extraction was done with reverse osmosis water at dierent temperature. Plant material to water ratio was maintained in the range of 0.02ÿ0.1. The extract from this column was pretreated with a ceramic tubular membrane (US Filter, Warren-dale, PA, USA) with a mean pore size of 0.35 mm. This membrane had a surface area of 0.005 m2and was oper-ated at a trans-membrane pressure of 104 kPa. Permeate from above was treated with an ulra®ltration (UF) mem-brane (Liumar Technologies, Ottawa, Canada) in dia®l-tration mode. This UF membrane had a molecular weight cut-o (MWCO) rating of 2.5 kDa and was operated at a trans-membrane pressure of 440 kPa. Dia®ltration was discontinued once the majority of the extracted glycosides were washed out in the permeate. This permeate was washed of lower molecular weight impurities by a
Table 1
The eects of pH on stevioside and color extraction pH Stevioside (mg/l) Optical absorbency at 420 nm 2.0 8100 5.9 7.0 8000 9.0 9.0 7900 8.0
Fig. 1. Concentration of various sweeteners versus the cumulative extract volumes at dierent temperatures. RA and RC are Rebaudio-side A and RebaudioRebaudio-side C, respectively.
nano®ltration (NF) membrane (DurathermTM, Osmo-nics/Desalination, Minnetonka, MN, USA) that was operated at a trans-membrane pressure of 510 kPa in a dia®ltration mode while feed could be maintained up to a temperature of 80C. Finally, the NF membrane was operated in a concentration mode and retentate was recovered for further processing.
3. Results and discussion
Development of this process involves several distinct unit operations that will be discussed below.
Based on a number of preliminary experiments, it was found that dried leaves of 10±40 mm size were most suitable for column extraction. A ¯ow rate of 24±30 ml/
min was maintained through the glass column. A tem-perature range of 4±35C for extracting water of varying pH was investigated. The eects of pH of extracting water are shown in Table 1. It is clear from the data that amount of stevioside in the extract remained constant while the amount of color components extracted were lower at lower pH. It was also observed that amount of extracted sweeteners as well as color components were higher at higher temperature. The rates of extraction of various sweeteners were relatively constant whereas at higher temperature the extraction rates were initially higher then gradually achieved a constant value. (Fig. 1). In order to remove suspended ®ne particles and very large molecular weight components, a laboratory scale ceramic micro®ltration (MF) membranes with a mean pore radius of 0.3±0.8 mm were found adequate. It was observed that about 80% of sweeteners permeated through these membranes. The rest of the sweeteners could be recovered by addition of water to concentrate followed by micro®ltration.
Fig. 2. Variation in ultra®ltration ¯ux with and without chemical pretreatment.
Table 2
Stevioside concentration and permeation rates for dierent dia®ltra-tion volumes in ultra®ltradia®ltra-tion step
Dia®ltration volume Stevioside concentration (g/l) Permeation rates (l mÿ2hÿ1) 0 1.5 35 2.0 0.75 47 3.0 0.25 64 4.0 NDa 65 a ND, not detected.
Fig. 3. A conceptual process ¯ow diagram for producing stevia sweetener concentrate. Hex-heat exchanger, DF, UF and NF are dia®ltration, ultra®ltration and nano®ltration, respectively.
The permeate stream from MF treatment contained most of the sweeteners and other natural products. This stream was treated with a selected ultra®ltration (UF) membrane. Addition of small amounts (less than 1% w/w) of ¯occulating agents improved the ¯ux at this stage as shown in Fig. 2. The UF membrane with a molecular weight cut-o rating of 2.5±3.0 kDa was used. It was found that about three dia®ltration volumes were sucient to permeate sweeteners through this membrane while larger molecular weight components were rejected (Table 2).
Permeate from UF stage contained sweeteners and very low molecular weight impurities. This stream was concentrated using a nano®ltartion membrane at higher temperatures. It was observed that in addition to con-centrating the sweeteners this treatment improved the taste pro®le of the product as determined by pre-liminary taste tastes and HPLC analysis. It was shown by the decline in the non-glycoside peaks in HPLC chromatograms that operation at higher temperature of 80C reduced the impurities to 55% of original while at 50C, these were reduced only to 89%.
Based on the above steps a conceptual ¯ow diagram of a process involving column extraction at a suitable tem-perature with multiple stage of membrane treatment is shown in Fig. 3. The process could provide a relatively high purity sweetener concentrate that could be further processed by ion exchange followed by spray drying to make high purity sweetener powder. Additional mem-brane units could be used to recycle the process water.
4. Conclusions
It was shown that extraction of sweeteners as well as other components was dependent on temperature of the
extracting water. The pH of extracting water did not aect the amount of extracted sweeteners, however, lower amounts of color components were extracted at a lower pH.
Pre-treatment with micro®ltration ceramic membranes was adequate. Addition of lime and or ¯occulating agent to ultra®ltration feed improved the ¯ux signi®cantly. A dia®ltration volume of 3 was found to be adequate for eluting sweeteners. The nano®ltration membranes were adequate for concentrating the sweeteners as well as removing lower molecular weight impurities. It was also observed that operation of nano®ltration membranes at higher temperatures was more eective in removing the impurities and consequently improving the taste pro®les of the sweeteners.
References
Brandle, J. (1999). Genetic control of rebaudioside A and C con-centration in leaves of the sweet herb, stevia rebaudiana. Canadian Journal of Plant Science, 79(1), 85±92.
Cramer, B., & Ikan, R. (1986). Sweet glycosides from the stevia plant. Chemistry in Britain, 22, 915±916.
Giovanetto, R. H. (1990). Method for the recovery of steviosides from plant raw material.US patent no. 4,892,938.
Kikuchi, H. (1985). Food Chemicals. Food Science (Japan), 85(3), 52. Kumar, S. (1986). Method for recovery of stevioside. US patent no.
4,599403.
Leung, A. Y., & Foster, S. (1996). Encyclopedia of common natural ingredients Ð used in food, drugs and cosmetics(2nd ed.). New York: John Wiley and Sons, Inc. (p. 478).
Matsushita, S., & Ikushige, T. (1979). Separation of natural sweet component from natural extract.US patent no. 4,1717430.
Sourirajan, S., & Matsuura, T. (1985). Reverse osmosis/ultra®ltration process principle. NRCC no. 24188 (pp. 79±81).
Zhang, S. Q., Kutowy, O., Kumar, A., & Malcolm, I. (1997). A laboratory study of poultry abattoir wastewater treatment by mem-brane technology. Canadian Agricultural Engineering, 39(2), 99±105. 620 S.Q. Zhang et al. / Food Research International 33 (2000) 617±620
